Carissa Capuano College of William and Mary for the G 0 Collaboration Hall C Users Meeting January 14, 2012

Parity-Violating Asymmetry in Electroproduction of the :

Inelastic Electron and Pion Results from the G0 Experiment at Backward Angle

Carissa CapuanoCollege of William and Mary

for the G0 Collaboration

Hall C Users MeetingJanuary 14, 2012

C. Capuano ~ College of W&M 2

G0 Inelastics: Overview• Purpose:

♦ Measurement of axial transition form factor, • 0.2 (GeV/c)2 < Q2 < 0.5 (GeV/c)2

• What does tell us? ♦ → Axial elastic form factor for N

• How is the spin distributed?♦ → Axial transition form factor for N →Δ

• How is the spin redistributed during transition?

• What do we measure?♦ Parity violating asymmetry Ainel

• Allows a direct measure of the axial (intrinsic spin) response during N →Δ

• Accessing :♦ Previous Measurements: Charged current process (W± exchange)

• Both quark flavor change and spin flip♦ G0 N-Δ Measurement: Neutral current process (Z0 exchange )

• Quark spin flip only

→ First measurement in neutral current sector

January 14, 2012


PV Vector Hadron Vertex

Resonant: = 2(1-2sin2 θW) ≈ 1Non-Resonant:

PV Axial Vector Hadron Vertex

Resonant: = 2(1-4sin2θW) F(Q2,s)Non-resonant: Neglected

Inelastic Asymmetry FormalismM.J. Musolf et al. Phys. Rept. 239 (1994)

𝐴𝑖𝑛𝑒𝑙=−𝐺𝐹𝑄

2

4𝜋𝛼 √2 [ Δ(1)𝜋 +Δ(2)

𝜋 +Δ(3)𝜋 ]

𝑭 (𝑸𝟐 ,𝒔 )=𝑬+𝑬 ′

𝑴𝑯𝑬𝑴 (𝑸𝟐 ,𝜽)𝑮𝑵 𝜟

𝑨 (𝑸𝟐)

Axial Form

Factors

EM Form Factors

A VV A

January 14, 2012


Axial Electroweak Radiative Effects

Rewrite to include EW radiative effects: One-quark: Interactions between gauge boson and constituent quarks

♦ Corrections to SM couplings - well known• Can be calculated using info from PDG

♦ Calculated to be ~60%• For vector terms: ~1-1.5%

♦ Applied to theoretical inelastic asymmetry

Multi-quark: Interactions between quarks in the nucleon

♦ May be significant for axial term, but high theoretical uncertainty• Negligible for vector hadron terms

♦ Especially interesting at low Q2

January 14, 2012

𝑹𝑨𝚫=𝑹𝑨

𝟏𝒒+𝑹𝑨𝒎𝒖𝒍𝒕𝒊

See Zhu et al. PRD 65 (2001) 033001


Seigert Term, and Pion Asymmetry

At the Q2 → 0 limit, asymmetry may not vanish

♦ A large non-zero asymmetry could explain large asymmetries in hyperon decay

Size of will depend on the size of ♦ Low energy coupling constant characterizing the PV vertex♦ Given in terms of = 5 x 10-8

♦ Zhu et al. theorized a “reasonable range” of (1-100)• Corresponds to

Can be studied using G0 pion data from LD2 at 362 MeV♦ Measurement performed at Q2 = 0.003 (GeV/c)2

♦ is a linear combo of photo- ( and electroproduced ( pions

Will also be studied by Qweak using inelastic ep data♦ Measurement of Ainel at (GeV/c)2

January 14, 2012

Zhu et al. PRD 65 (2001) 033001


G0 Experimental Setup

Polarized Beam:♦ Longitudinally polarized

beam • Pb = 85%

Unpolarized Cryotarget:♦ LH2 or LD2

Detector System:♦ Scintillators:

• Two sets allow for kinematic separation of elastic and inelastic regions

– Cryostat Exit Detectors (CED)

– Focal Plane Detectors (FPD)♦ Cherenkov Detectors (CER):

• Allow us to distinguish between pions and electrons

♦ Measured events: Coincidences

• CED + FPD + CER fire→ electron

• CED + FPD fire (CER doesn’t fire)→ pion

e- beam target

CED + Cherenkov

FPD

Cutaway view of a single octant

Eight detector arrays like the one above are arranged symmetrically around the target

January 14, 2012


Data Analysis: Summary

Correct for beam and instrumentation♦ Dead time and randoms♦ Helicity correlated beam properties♦ Beam polarization♦ Transverse polarization

Correct for Backgrounds♦ Inelastic: Significant background fraction; dominated by elastic

radiative tail♦ Pions: Small background, big effect on asymmetry; dominated

by electron contamination

Correct for EM radiation & acceptance averaging♦ Inelastic hydrogen only!

Once all corrections are applied, can extract physics results from the measured asymmetries

January 14, 2012


Final Corrected Asymmetries

Inelastic Data: W = 1.18 GeV, Q2 = 0.34 (GeV/c)2

ADinel = -43.6 ± (14.6)stat ± (6.2)sys ppm

AHinel = -33.4 ± (5.3)stat ± (5.1)sys ppm

**Form factor determination will be for H result only

Pion Data: W = 1.22 GeV, Q2 = 0.0032 (GeV/c)2

A = -0.55 ± (1.03)stat + (0.37)sys ppmJanuary 14, 2012


Comparison: Measured Ainel vs. Theory

January 14, 2012

Inelastic hydrogen result: Compare to theoretical total asymmetry and individual components.


Extracting the Axial Form Factor,

First need to isolate

Assuming A1 and A2 are known,

From , extract

January 14, 2012

ppm

𝑮𝑵 𝜟𝑨 =− 𝑴

𝑬+𝑬 ′

𝟐𝝅𝜶 √𝟐𝑮𝑭𝑸

𝟐

𝑨𝟑

𝟐𝑯 𝑬𝑴 (𝑸𝟐 ,𝜽 ) [𝟏−𝟒𝐬𝐢𝐧𝟐 𝜽𝑾 ]

𝑮𝑵 𝚫𝑨 =−𝟎 .𝟎𝟓± (𝟎 .𝟑𝟓 )𝒔𝒕𝒂𝒕+ (𝟎 .𝟑𝟒 )𝒔𝒚𝒔+ (𝟎 .𝟎𝟔 )𝒕𝒉

(Theory: )

(Theory: ppm)

𝐴𝑖𝑛𝑒𝑙=𝐴1+𝐴2+𝐴3=−𝐺𝐹𝑄

2

4𝜋𝛼 √2 [ Δ(1)𝜋 +Δ(2)

𝜋 +Δ(3)𝜋 ]


Extracting the coupling constant,

First need to find photoproduction asymmetry,

Use input from theory and simulation to isolate

• estimate as

From , extract January 14, 2012

(Theory: )


Final Summary

• Measurement: PV asymmetry in electroproduction of the ♦ E = 687 MeV, D target – Determine Ainel

♦ E = 687 MeV, H target – Determine Ainel and form factor

♦ E = 362 MeV, D target – Determine and

• Results:♦ Inelastic Data:

• First measurement using neutral current process• Form factor found to be consistent with theory, but large error

♦ Pion Data:• Resulted in ±25 bound on → |A(Q2=0)| < 2 ppm

• Publications:

♦ Pion Result: arXiv:1112.1720v1 [nucl-ex] (submitted to PRL)

♦ Inelastic Result: Coming soon…

January 14, 2012

Backup Slides


Computing the Axial Component

• Requires neutral weak axial and vector form factors

♦ CVC Hypothesis: Replace vector with EM form factors• EM FF’s well known

♦ Isospin Rotation: Replace axial with CC axial form factors• CC FF’s determined from neutrino data

• Basic Form: Adler Parameterization

January 14, 2012

𝑭 (𝑸𝟐 ,𝒔 )=𝑬+𝑬 ′

𝑴𝑯𝑬𝑴 (𝑸𝟐 ,𝜽)𝑮𝑵 𝜟

𝑨 (𝑸𝟐)

Vector:

Axial:

Depend on the Adler form factors,

Unknown

Dipole FormExtra Q2

Dependence


Axial Electroweak Radiative Corrections

Rewite to include effects:

January 14, 2012

𝑹𝑨𝚫=𝑹𝑨

𝒆𝒘𝒌+𝑹𝑨𝑺𝒊𝒆𝒈𝒆𝒓𝒕+𝑹𝑨

𝒂𝒏𝒂𝒑𝒐𝒍𝒆+𝑹𝑨𝒅−𝒘𝒂𝒗𝒆+…

tree-level

PV γNΔ vertex

PV NΔ vertex

1-quark

Negligible

Inelastic measurement: Anapole may contribute

~0.3ppm but high theoretical uncertainty

→Multiquark corrections neglected

60% effect

Pion Measurement:Siegert term dominates, size

depends on coupling constant

Zhu et al. PRD 65 (2001) 033001


Axial Multi-quark EW Radiative Effects

January 14, 2012

Note: Figure taken from Zhu

et al., not at exact G0

kinematics

Inelastic:Q2 = 0.34 GeV2

A3

anapole

Siegert ()

d-wave

Zhu et al. PRD 65 (2001) 033001

Pion:Q2 = 0.003

GeV2

r i =

Ai /

Ato

t


Superconducting Magnet (SMS)

Detectors:Ferris Wheel

(FPDs)Detectors:

Mini-Ferris wheel(CEDs+Cherenkov)

Target Service Module

G0 Beam Monitoring

“Front” View:

The G0 Experiment in Hall C

January 14, 2012


CED

H 687 Electron Yield (Octant 2)

Detector Acceptance and Yields

D 687 Electron Yield (Octant2)

CED

FPD

CED

inel

astic

s

FPD

elas

tics

inel

astic

s

elastics

January 14, 2012

** Similar matrices exist for pion data


Detector Acceptance and Yields

January 14, 2012

D 362 Pion Yield (Octant Average)


Data Summary

Two Targets: Needed for elastic measurement♦ Elastic asymmetry contains 3 form factors♦ Forward H + Backward H + D allows full separation

Two Energies: Allows for elastic result at two Q2 points

Only high energy run periods useful for inelastic measurementOnly low energy D run period used for pion measurement

January 14, 2012

Date Target Ebeam (MeV) Ibeam(A) Charge(C) # Runs

Apr ’06 H 685.6 60 16.3 100

Sep-Oct ’06 H 684.9 60 97.1 548

Nov-Dec ’06 D 689.6 20 32.8 532

Mar ’07 D 689.4 17 17.3 332

Jul-Aug ’06 H 361.9 60 78.0 475

Jan-Feb ‘07 D 363.1 35 67.4 649


Scaler Counting Correction

Symptom: Tails on the yield♦ D 362 data most affected♦ Rate dependent → Impact on inelastic cells minimal

Problem: Bad MPS counts in NA octants♦ NA coincidence boards did not have a

minimum output width♦ Scaler boards didn’t properly handle

consecutive short pulses→ Two effects combined lead to dropped bits

Solution: Program a minimum output width of 10ns

→ Problem diagnosed and corrected during experimental run

Correction: Remove QRTs with bad MPSs♦ Events outside ±5σ window removed from

averaging

Impact: Tails removed w/o negatively impacting unaffected data

♦ Bad MPS Uncorrelated across cells→ Correction results in 1% of events cut in D

362 run period→ 0.1% in all othersJanuary 14, 2012

raw

corre

cte

d

Asymmetry

Yield

FR NA

Yield


Rate Corrections: Inelastic Data

Dead Time: ♦ Real events missed while electronics processed previous events → adds events

Accounts for components of the CED and FPD electronics Does not include Cerenkov DT

Contamination: ♦ Misidentified particles → adds & subtracts events

Cerenkov dead time – e in matrix Cerenkov randoms – in e matrix

Randoms: ♦ Random CED·FPD coincidences → subtracts events

Only applied to the pion matrix

Overall impact of rate corrections on asymmetry → net effect

Uncertainty:♦ False asymmetry from residual DT → negligible

♦ False asymmetry from CED·FPD·CER randoms

→ Bound inelastic locus uncertainty using information from elastic analysisJanuary 14, 2012

dA =

Error ~10%of correction


Helicity Correlated Beam Properties

• Correct for false asymmetry due to changes in…♦ Beam position in x or y

direction♦ Beam angle in x or y direction♦ Beam Current♦ Beam Energy

• Size of correction determined by beam quality♦ Specifications given to ensure

sufficient precision

𝐴 𝑓𝑎𝑙𝑠𝑒=∑ 12𝑌

𝜕𝑌𝜕𝑃 𝑖

∆𝑃𝑖

Spec Actual

40 -19 3

40 -17 2

4 -0.8 0.2

4 0.0 0.1

34 2.5 0.5

2 0.09 0.08

January 14, 2012

|Afalse|< 0.3 ppm ⟹

Transverse Asymmetry Correction: Inelastic Data

Correct for false asymmetry arising from transverse beam:

• Impact depends on…♦ Magnitude of transverse asymmetry,

♦ Determined through direct measurement

♦ Physical misalignment in detector system, ♦ Sinusoidal octant dependence

→ Should cancel in a symmetrical detector system

♦ Degree of transverse polarization, ♦ Determined from LUMI data

• Computed upper bound, found to be small (< 0.05 ppm)♦ Consistent with elastic locus results

→ No correction applied, treated as uncertainty

𝑨𝑻𝒄𝒐𝒓𝒓=𝑨𝑻𝑴𝒅𝒆𝒕

𝑷𝑻

𝑷

TransverseLongitudinal

-1 → 1 -20 → 20

Difficult to quantify


Beam Polarization

• Polarimeter: Measure an asymmetry using Møller scattering

♦ Polarized iron target♦ θ = 90°

• Measurements performed periodically throughout the experimental run

♦ Pb stable throughout

⟹𝑃687=¿January 14, 2012


Background Correction: Inelastic Data• Contributing processes:

♦ Electrons from inelastic e-p(d) scattering♦ Electrons from elastic e-p(d) scattering ♦ Electrons from 0 decay

♦ Electrons scattered from Al target windows

♦ Contamination from - (D target only)

• Fitting: Scale Yield vs. FPD for each CED♦ Before fitting, subtract - contamination and target window yield ♦ Scale the remaining contributions independently to fit the data

• Fit Requirements:♦ Fit across all octants - forces all to have the same scale factor ♦ Require scale factors to vary smoothly across CEDs

)()()()( 0210 fpdYPfpdYPfpdYPfpdY inelelfit

“Empty target” data **

GEANT Simulation

** Gas target data scaled to remove the gas contribution and to account for the kinematic differences in the liquid and gas target

January 14, 2012

Pion data analysis


Correcting the Asymmetry:♦ Extract Ainel from Ameas by subtracting off backgrounds

♦ High backgrounds: ~50% for H, ~65% for D

Background Asymmetries:♦ Elastic Electrons

• Use Ael measured by G0

• Dominated by radiative tail → Use simulation to determine a scale factor

♦ Target windows• Dominated by inelastic events• Ainel

al is unknown, but can use measured D asymmetry♦ Pion related: Misidentified - and electrons from 0 decay

• A measured by G0

January 14, 2012

𝑨𝒊𝒏𝒆𝒍=𝑨𝒎𝒆𝒂𝒔−∑ 𝒇 𝒊

𝒃𝒈𝑨𝒊𝒃𝒈

𝟏−∑ 𝒇 𝒊𝒃𝒈

Background Correction: Application

Impact on Asymmetry:

26% change for H, 40% change for D

Impact on Uncertainty:

Significant increase - more than doubled


Background Correction: Pion Data

• Method: Use time of flight spectra from 31MHz pulsed beam

• Primary source: Misidentified electrons

• Particle ID: Use ToF cuts to define true e and rates, compare to data to get efficiency

January 14, 2012

• Backgrounds: ♦ 2.6% electrons

scattered from target liquid

♦ 2% Al target windows can be ignoredD target!

• Apply Correction: Same procedure as inelastics

C. Capuano ~ College of W&M 29January 14, 2012

Correction A_inel s_tot s_stat s_sys s_cor dA_corr

Raw -14.11 2.62 2.62 0.00 --- ---

Scalar Counting Prob.

-14.06 2.62 2.62 0.00 0.00 +0.05

Rate Corrections -26.66 5.99 5.87 1.20 1.20 -12.6

Linear Regression -26.41 6.01 5.88 1.23 0.25 +0.25

Beam Polarization -31.07 7.04 6.92 1.30 0.43 -4.66

Transverse -31.07 7.04 6.92 1.30 0.02 ---

Backgrounds -43.57 15.91 14.64 6.23 5.52 -12.5

ADinel = -43.57 ± 15.9 ppm

All values in ppm

D 687 Inelastics: Summary of Corrections & Error


Correction A_inel s_tot s_stat s_sys s_cor dA_corr

Raw -20.23 2.00 2.00 0.00 --- ---

Scalar Counting Prob -20.00 1.99 1.99 0.00 0.00 +0.23

Rate Corrections -22.17 2.26 2.25 0.16 0.16 -2.17

Linear Regression -22.33 2.25 2.24 0.23 0.16 -0.16

Beam Polarization -26.27 2.64 2.64 0.43 0.36 -3.91

Transverse -26.27 2.64 2.64 0.43 0.03 ---

Backgrounds -33.60 7.36 5.30 5.10 4.93 -7.33

EM Radiation -33.99 7.36 5.30 5.10 0.20 -0.39

Acceptance Avg. -33.44 7.36 5.30 5.13 0.55 +0.55

AHinel = -33.44 ± 7.4 ppm

All values in ppm

January 14, 2012

H 687 Inelastics: Summary of Corrections & Error

C. Capuano ~ College of W&M 31January 14, 2012

Correction A_pi s_stat s_cor

Raw -0.17 ---

Scalar Counting Prob. -0.17 0.75 0.00

Rate Corrections -0.54 0.78 0.26

Linear Regression -0.52 0.78 0.21

Backgrounds -0.22 0.88 0.12

Transverse -0.45 0.89 0.08

Polarization -0.55 1.03 0.01

A = -0.55 ± 1.1 ppm

All values in ppm

D 362 Pions: Summary of Corrections & Error

Documents

Carissa Capuano College of William and Mary for the G 0 Collaboration Hall C Users Meeting January 14, 2012